entry_SYSCALL_64_after_hwframe+0x77/0x7f loop2: detected capacity change from 0 to 15 loop1: detected capacity change from 0 to 15 ------------[ cut here ]------------ percpu ref (css_release) <= 0 (-1) after switching to atomic WARNING: lib/percpu-refcount.c:197 at percpu_ref_switch_to_atomic_rcu+0x3cc/0x480, CPU#0: ksoftirqd/0/14 Modules linked in: CPU: 0 UID: 0 PID: 14 Comm: ksoftirqd/0 Tainted: G W 6.17.0-rc3-next-20250829 #1 PREEMPT(voluntary) Tainted: [W]=WARN Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 RIP: 0010:percpu_ref_switch_to_atomic_rcu+0x3cc/0x480 Code: 00 00 00 00 00 fc ff df 48 89 f9 48 c1 e9 03 80 3c 01 00 0f 85 9e 00 00 00 49 8b 75 e8 48 c7 c7 80 97 e2 84 e8 75 c5 e9 fe 90 <0f> 0b 90 90 e9 2b ff ff ff e8 f6 de 5f ff e9 9e fe ff ff e8 7c df RSP: 0018:ffff888009617c70 EFLAGS: 00010282 RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffffffff8139de70 RDX: ffff888009601b80 RSI: ffffffff8139de7e RDI: 0000000000000001 RBP: 7fffffffffffffff R08: 0000000000000001 R09: ffffed100d9c4801 R10: 0000000000000000 R11: 0000000000000001 R12: ffff88800c840500 R13: ffff88800c840520 R14: 0000000000000002 R15: 0000000000000003 FS: 0000000000000000(0000) GS:ffff8880e55dd000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007ffe17911000 CR3: 000000000fde0000 CR4: 0000000000350ef0 Call Trace: rcu_core+0x7c8/0x1800 handle_softirqs+0x1b1/0x770 run_ksoftirqd+0x2e/0x60 smpboot_thread_fn+0x41d/0x9d0 kthread+0x3c8/0x740 ret_from_fork+0x34b/0x430 ret_from_fork_asm+0x1a/0x30 irq event stamp: 10022912 hardirqs last enabled at (10022920): [] __up_console_sem+0x78/0x80 hardirqs last disabled at (10022929): [] __up_console_sem+0x5d/0x80 softirqs last enabled at (10022490): [] handle_softirqs+0x50c/0x770 softirqs last disabled at (10022495): [] run_ksoftirqd+0x2e/0x60 ---[ end trace 0000000000000000 ]--- percpu_ref_switch_to_atomic_rcu: percpu_ref_switch_to_atomic_rcu(): percpu_ref underflow slab kmalloc-64 start ffff88800c840500 pointer offset 0 size 64 percpu_ref_switch_to_atomic_rcu: percpu_ref_switch_to_atomic_rcu(): percpu_ref underflow slab kmalloc-64 start ffff88800d336500 pointer offset 0 size 64 ================================================================== BUG: KASAN: slab-use-after-free in mem_cgroup_wb_domain+0x72/0x80 Read of size 8 at addr ffff88801561e988 by task kworker/u9:6/3067 CPU: 0 UID: 0 PID: 3067 Comm: kworker/u9:6 Tainted: G W 6.17.0-rc3-next-20250829 #1 PREEMPT(voluntary) Tainted: [W]=WARN Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 Workqueue: writeback wb_workfn (flush-8:0) Call Trace: dump_stack_lvl+0xca/0x120 print_report+0xcb/0x610 kasan_report+0xca/0x100 mem_cgroup_wb_domain+0x72/0x80 wb_over_bg_thresh+0xbb/0x180 wb_workfn+0x4b1/0xb50 process_one_work+0x8e1/0x19c0 worker_thread+0x67e/0xe90 kthread+0x3c8/0x740 ret_from_fork+0x34b/0x430 ret_from_fork_asm+0x1a/0x30 Allocated by task 1: kasan_save_stack+0x24/0x50 kasan_save_track+0x14/0x30 __kasan_slab_alloc+0x59/0x70 kmem_cache_alloc_noprof+0x205/0x690 mem_cgroup_css_alloc+0x57/0x15e0 cgroup_apply_control_enable+0x446/0x9f0 cgroup_mkdir+0x86e/0x1110 kernfs_iop_mkdir+0x111/0x190 vfs_mkdir+0x59a/0x8d0 do_mkdirat+0x19f/0x3d0 __x64_sys_mkdir+0xf3/0x140 do_syscall_64+0xbf/0x360 entry_SYSCALL_64_after_hwframe+0x77/0x7f Freed by task 9: kasan_save_stack+0x24/0x50 kasan_save_track+0x14/0x30 __kasan_save_free_info+0x3a/0x60 __kasan_slab_free+0x3f/0x50 kfree+0x281/0x550 css_free_rwork_fn+0x125/0x1110 process_one_work+0x8e1/0x19c0 worker_thread+0x67e/0xe90 kthread+0x3c8/0x740 ret_from_fork+0x34b/0x430 ret_from_fork_asm+0x1a/0x30 Last potentially related work creation: kasan_save_stack+0x24/0x50 kasan_record_aux_stack+0x89/0xa0 __queue_work+0xa3b/0x1240 rcu_work_rcufn+0x5c/0x90 rcu_core+0x7c8/0x1800 handle_softirqs+0x1b1/0x770 __irq_exit_rcu+0xc4/0x100 irq_exit_rcu+0x9/0x20 sysvec_apic_timer_interrupt+0x70/0x80 asm_sysvec_apic_timer_interrupt+0x1a/0x20 Second to last potentially related work creation: kasan_save_stack+0x24/0x50 kasan_record_aux_stack+0x89/0xa0 __call_rcu_common.constprop.0+0x70/0x960 queue_rcu_work+0xa9/0xe0 process_one_work+0x8e1/0x19c0 worker_thread+0x67e/0xe90 kthread+0x3c8/0x740 ret_from_fork+0x34b/0x430 ret_from_fork_asm+0x1a/0x30 The buggy address belongs to the object at ffff88801561e880 which belongs to the cache mem_cgroup of size 2248 The buggy address is located 264 bytes inside of freed 2248-byte region [ffff88801561e880, ffff88801561f148) The buggy address belongs to the physical page: page: refcount:0 mapcount:0 mapping:0000000000000000 index:0xffff88801561e880 pfn:0x15618 head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0 flags: 0x100000000000040(head|node=0|zone=1) page_type: f5(slab) raw: 0100000000000040 ffff888009425780 dead000000000122 0000000000000000 raw: ffff88801561e880 00000000800d000b 00000000f5000000 0000000000000000 head: 0100000000000040 ffff888009425780 dead000000000122 0000000000000000 head: ffff88801561e880 00000000800d000b 00000000f5000000 0000000000000000 head: 0100000000000003 ffffea0000558601 00000000ffffffff 00000000ffffffff head: 0000000000000000 0000000000000000 00000000ffffffff 0000000000000008 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff88801561e880: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88801561e900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb >ffff88801561e980: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ^ ffff88801561ea00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ffff88801561ea80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb ================================================================== loop2: detected capacity change from 0 to 15 loop3: detected capacity change from 0 to 15 loop1: detected capacity change from 0 to 15 kmemleak: Found object by alias at 0x607f1a638c34 CPU: 0 UID: 0 PID: 25931 Comm: syz-executor.2 Tainted: G B W 6.17.0-rc3-next-20250829 #1 PREEMPT(voluntary) Tainted: [B]=BAD_PAGE, [W]=WARN Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 Call Trace: dump_stack_lvl+0xca/0x120 __lookup_object+0x94/0xb0 delete_object_full+0x27/0x70 free_percpu+0x30/0x1160 futex_hash_free+0x38/0xc0 mmput+0x2d3/0x390 do_exit+0x79d/0x2970 do_group_exit+0xd3/0x2a0 __x64_sys_exit_group+0x3e/0x50 x64_sys_call+0x18c5/0x18d0 do_syscall_64+0xbf/0x360 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f447c6eeb19 Code: Unable to access opcode bytes at 0x7f447c6eeaef. RSP: 002b:00007fff307ecfb8 EFLAGS: 00000246 ORIG_RAX: 00000000000000e7 RAX: ffffffffffffffda RBX: 000000000000001e RCX: 00007f447c6eeb19 RDX: 00007f447c6a172b RSI: ffffffffffffffbc RDI: 0000000000000000 RBP: 0000000000000000 R08: 0000001b2d224764 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001 R13: 0000000000000000 R14: 0000000000000001 R15: 00007fff307ed0a0 kmemleak: Object (percpu) 0x607f1a638c30 (size 8): kmemleak: comm "syz-executor.0", pid 25932, jiffies 4295577284 kmemleak: min_count = 1 kmemleak: count = 0 kmemleak: flags = 0x21 kmemleak: checksum = 0 kmemleak: backtrace: pcpu_alloc_noprof+0x87a/0x1170 alloc_vfsmnt+0x135/0x6e0 clone_mnt+0x6c/0xb70 __do_loopback+0x438/0x520 path_mount+0x19b4/0x1dd0 __x64_sys_mount+0x27b/0x300 do_syscall_64+0xbf/0x360 entry_SYSCALL_64_after_hwframe+0x77/0x7f loop6: detected capacity change from 0 to 15 loop0: detected capacity change from 0 to 15 loop4: detected capacity change from 0 to 15